University of Sheffield at TRECVID 2006 High-level Feature Extraction

We present our approach to TRECVID 2006, high-level feature extraction task. We submitted one run with type ‘A’, annotating all required 39 features. The approach was based on textual information extracted from speech recogniser and machine translation outputs. They were aligned with shots and associated with highlevel feature references. A list of significant words was created for each feature, and it was in turn utilised for identification of a feature during the evaluation. In this notebook paper, we describe the approach and the results we obtained. We also describe the problems we encountered during the system development, some of which were critical to the system performance.